1. Technical Field
The present invention generally relates to computer network systems and in particular to directory servers in computer network systems. Still more particularly, the present invention relates to the determination of master directory servers in computer network systems.
2. Description of the Related Art
Conventional directory servers allow different servers to manage different branches of a directory tree. However, the allocation of branches to servers is a static or manual decision. This means that an implementer needs to make a conscious choice about which server will control particular branches. If a particular branch is accessed more by users on a server which does not control that branch, additional time is required to forward the users' requests to the appropriate server compared to the time taken to fulfill requests for branches the server actually controls.
The additional time consumed forwarding numerous requests to another server results in inefficiencies with the current method of static or manual allocation of a master server. Unfortunately, the server that controls particular branches at implementation remains in control of these branches indefinitely or until an implementer makes a manual change of the branch-master server allocation. Considering the inefficiencies in the system of manually designating master servers, the present invention appreciates that a solution which places branches under the control of the server at which they are requested most would increase the efficiency of the system of server directories.